An Enhanced Laryngeal Tissue Classification Model Using Deep Learning Techniques

The current research introduces a deep learning methodology utilizing Convolutional Neural Networks (CNNs) for the categorization of laryngeal tissues. In the proposed approach six dense layers was added after VGG-19 and was evaluated on a corpus of laryngeal images and demonstrated proficiency in categorizing four distinct varieties of laryngeal tissues, namely, normal, exhibiting hypertrophic vessels, leukoplakia, and intrapapillary capillary loops. The corpus of laryngeal images used for evaluation was collected from a diverse range of patients. The results of the study indicate that the proposed methodology has attained significant accuracy in categorizing diverse laryngeal tissues, thereby showcasing its viability for implementation in clinical settings. The approach that has been proposed exhibits potential as a method for the automated categorization of laryngeal tissues. This has the potential to be of assistance in the identification and management of laryngeal problems. The proposed methodology could also be adapted for use in other medical image classification tasks, such as the classification of skin lesions or tumors.